Research On State Of Charge Estimation Of LiFePO₄ Battery

New energy vehicles have received strong support from the government.As the power battery for new energy vehicles,lithium iron phosphate battery has attracted the attention of many researchers.However,the development of new energy vehicles is constrained by lithium iron phosphate batteries,so it is necessary to study its performance and state of charge(SOC)to develop a better battery management system(BMS).In this paper,the performance of lithium iron phosphate battery is studied first,including its structure and operating principle.The battery charging and discharging test results are combined with mathematical formulas to introduce the factors that affect battery capacity and SOC.According to the battery test standards at home and abroad and the existing battery test platform,various battery test schemes of this subject were formulated.The charging and discharging rate characteristics of the battery,the temperature characteristics of the battery,as well as the relationship between the open circuit voltage of the battery and the SOC,were analyzed through experimental results.Subsequently,among the many battery modeling methods,Thevenin equivalent circuit model was selected as the model of the lithium iron phosphate battery studied in this paper,and the parameters were identified by recursive least squares method.In order to make the model universally applicable,14 sets of different single-cell pulse discharge data were selected and segmented for parameters identification.After the simulation of the acquired model in the Simulink,the results show that the parameters are accurate and can meet the actual calculation requirements.Based on the existing model,the Kalman filter algorithm is used to estimate the SOC of lithium iron phosphate battery.Since the commonly used extended Kalman filter algorithm still has some drawbacks,it is necessary to use the adaptive extended Kalman filter algorithm to correct the noise.After the battery SOC estimation algorithm based on adaptive extended Kalman filter is proposed,Simulink is used to compare the two algorithms with the amperetime integration method.The feasibility of the two algorithms and the advantages of the adaptive extended Kalman filter algorithm are proved.Finally,a battery monitoring system based on LabVIEW is designed,which includes the hardware experimental platform based on MICROCHIP's DSPIC30F6012 A chip and the host computer software MPLAB X IDE and LabVIEW.The hardware platform implements the functions of data acquisition,processing and transmission by using various modules.After the data is transmitted to the host computer,it is received and processed in the LabVIEW and displayed in the graphical user interface.The battery pack constant current discharge test indicates that the battery monitoring system can realize real-time monitoring of the information of each cell in the battery pack.